Bruxism is a masticatory muscle activity characterized by high prevalence, widespread complications, and serious consequences but without specific guidelines for its diagnosis and treatment. Although occlusal force-based biofeedback therapy is proven to be safe, effective, and with few side effects in improving bruxism, its mechanism and key technologies remain unclear. The purpose of this study was to research a real-time, quantitative, intelligent, and precise force-based biofeedback detection device based on artificial intelligence (AI) algorithms for the diagnosis and treatment of bruxism. Stress sensors were integrated and embedded into a resin-based occlusion stabilization splint by using a layering technique (sandwich method). The sensor system mainly consisted of a pressure signal acquisition module, a main control module, and a server terminal. A machine learning algorithm was leveraged for occlusal force data processing and parameter configuration. This study implemented a sensor prototype system from scratch to fully evaluate each component of the intelligent splint. Experiment results showed reasonable parameter metrics for the sensors system and demonstrated the feasibility of the proposed scheme for bruxism treatment. The intelligent occlusion stabilization splint with a stress sensor system is a promising approach to bruxism diagnosis and treatment.Sensors 2020, 20, 89 2 of 15 discussion summary, the existing assessment of bruxism could also be classified into three main aspects: (1) noninstrumental approaches, (2) instrumental approaches, and (3) cut-off points grading [6]. Self-report and clinical examination are considered as noninstrumental approaches, which are also the primary choices in the clinical assessment of bruxism. However, their reliability and validity need further improvement [6]. EMG may provide good evidence of both sleep and awake bruxism, but there is also a risk of overestimating the number of true SB events [7]. PSG could be regarded as a reference standard for SB assessment; however, it is expensive and time-consuming [2]. To date, the reliability and validity of all the common techniques remain debatable, and consensus has yet to be established regarding the best method to diagnose bruxism. Therefore, exploring some new methods of bruxism diagnosis and management is a necessary and meaningful research topic.The use of biofeedback technologies (electrical, auditory, vibratory stimulus, etc.) as behavioral techniques of bruxism diagnosis and treatment has been considered a promising approach in both clinical and scientific fields in recent years [8]. Various biofeedback modalities have been reported in previous papers, and most of these are based on EMG recordings, except two studies that described force-based devices [8,9]. An intra-splint force detector (ISFD) for SB force detection was described in 2003 [10]. A detailed description of this SB inhibition system has been presented in a recently published paper [11]. This inhibition system consisted of ISFD, vibration, and co...

No abstract

Pruning can remove redundant parameters and structures of Deep Neural Networks (DNNs) to reduce inference time and memory overhead. As an important component of neural networks, the feature map (FM) has stated to be adopted for network pruning. However, the majority of FM-based pruning methods do not fully investigate effective knowledge in the FM for pruning. In addition, it is challenging to design a robust pruning criterion with a small number of images and achieve parallel pruning due to the variability of FMs. In this paper, we propose Adaptive Knowledge Extraction for Channel Pruning (AKECP), which can compress the network fast and efficiently. In AKECP, we first investigate the characteristics of FMs and extract effective knowledge with an adaptive scheme. Secondly, we formulate the effective knowledge of FMs to measure the importance of corresponding network channels. Thirdly, thanks to the effective knowledge extraction, AKECP can efficiently and simultaneously prune all the layers with extremely few or even one image. Experimental results show that our method can compress various networks on different datasets without introducing additional constraints, and it has advanced the state-of-the-arts. Notably, for ResNet-110 on CIFAR-10, AKECP achieves 59.9% of parameters and 59.8% of FLOPs reduction with negligible accuracy loss. For ResNet-50 on ImageNet, AKECP saves 40.5% of memory footprint and reduces 44.1% of FLOPs with only 0.32% of Top-1 accuracy drop. The code is available at https://github.com/zhnxjtu/AKECP. CCS CONCEPTS• Computing methodologies → Computer vision.

No abstract

No abstract